Classification model of packed malware with attention mechanism

Songyuan Ma; Wanping Liu

doi:10.1117/12.2683184

26 June 2023 Classification model of packed malware with attention mechanism

Songyuan Ma, Wanping Liu

Proceedings Volume 12714, International Conference on Computer Network Security and Software Engineering (CNSSE 2023); 1271410 (2023) https://doi.org/10.1117/12.2683184
Event: Third International Conference on Computer Network Security and Software Engineering (CNSSE 2023), 2023, Sanya, China

Abstract

In recent years, deep learning has been applied to the field of malware detection to improve the detection accuracy. However, many malware detection models based on static features and visualization methods do not consider the impact of code confusion, and the detection rate of packed malware is low. In order to solve this problem, this paper proposes a model SE-MHSA combining channel attention and multi head attention mechanism. With ResNet as the backbone network, we combine the SE module and MHSA module to extract channel, local and global features, and weaken the interference caused by code obfuscation by fusing multi-level features, so that the model can capture the correlation information of each partial feature and improve the detection capability of the model. The experiment is carried out on the public data set Malimg, and the accuracy is 99.6%. The experiment is carried out on the packed data set Virushare-Packed, and the accuracy is 95%. Compared with other models, this model achieves better results, and verifies the generalization and anti-confusion ability of the proposed model.

Citation Download Citation

Songyuan Ma and Wanping Liu "Classification model of packed malware with attention mechanism", Proc. SPIE 12714, International Conference on Computer Network Security and Software Engineering (CNSSE 2023), 1271410 (26 June 2023); https://doi.org/10.1117/12.2683184

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available